Merge 7e51ac7712d122beff12d2ef8483b1951f325747 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',
-        'W9HJZKktxuYoDTqW', 'U48vOZHaeYS6vO',
+        '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',
+    ),
+    (
+        '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/atresplayer.py

@@ -6,14 +6,19 @@ import re
 from .common import InfoExtractor
 from ..compat import compat_HTTPError
 from ..utils import (
+    base_url,
     ExtractorError,
     int_or_none,
     urlencode_postdata,
+    urljoin,
+    xpath_element,
+    xpath_text,
+    xpath_with_ns,
 )
 
 
 class AtresPlayerIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?atresplayer\.com/[^/]+/[^/]+/[^/]+/[^/]+/(?P<display_id>.+?)_(?P<id>[0-9a-f]{24})'
+    _VALID_URL = r'https?://(?:www\.)?atresplayer\.com/[^/]+/[^/]+/[^/]+(?:/[^/]+)?/(?P<display_id>.+?)_(?P<id>[0-9a-f]{24})'
     _NETRC_MACHINE = 'atresplayer'
     _TESTS = [
         {
@@ -39,7 +44,6 @@ class AtresPlayerIE(InfoExtractor):
             'only_matching': True,
         },
     ]
-    _API_BASE = 'https://api.atresplayer.com/'
 
     def _real_initialize(self):
         self._login()
@@ -49,6 +53,8 @@ class AtresPlayerIE(InfoExtractor):
             error = self._parse_json(e.cause.read(), None)
             if error.get('error') == 'required_registered':
                 self.raise_login_required()
+            if error.get('error') == 'invalid_request':
+                raise ExtractorError('Authentication failed', expected=True)
             raise ExtractorError(error['error_description'], expected=True)
         raise
 
@@ -57,35 +63,65 @@ class AtresPlayerIE(InfoExtractor):
         if username is None:
             return
 
-        self._request_webpage(
-            self._API_BASE + 'login', None, 'Downloading login page')
-
         try:
-            target_url = self._download_json(
-                'https://account.atresmedia.com/api/login', None,
+            self._download_webpage(
+                'https://account.atresplayer.com/auth/v1/login', None,
                 'Logging in', headers={
                     'Content-Type': 'application/x-www-form-urlencoded'
                 }, data=urlencode_postdata({
                     'username': username,
                     'password': password,
-                }))['targetUrl']
+                }))
         except ExtractorError as e:
             self._handle_error(e, 400)
 
-        self._request_webpage(target_url, None, 'Following Target URL')
+    def _get_mpd_subtitles(self, mpd_xml, mpd_url):
+        subs = {}
+
+        def _add_ns(name):
+            return xpath_with_ns(name, {
+                'mpd': 'urn:mpeg:dash:schema:mpd:2011'
+            })
+
+        def _is_mime_type(node, mime_type):
+            return node.attrib.get('mimeType') == mime_type
+
+        text_nodes = mpd_xml.findall(
+            _add_ns('mpd:Period/mpd:AdaptationSet[@contentType="text"]'))
+        for adaptation_set in text_nodes:
+            lang = adaptation_set.attrib['lang']
+            representation = xpath_element(adaptation_set, _add_ns('mpd:Representation'))
+            subs_url = xpath_text(representation, _add_ns('mpd:BaseURL'))
+            if subs_url and (_is_mime_type(adaptation_set, 'text/vtt') or _is_mime_type(representation, 'text/vtt')):
+                subs.update({lang: [{
+                    'ext': 'vtt',
+                    'url': urljoin(mpd_url, subs_url),
+                }]})
+        return subs
 
     def _real_extract(self, url):
         display_id, video_id = re.match(self._VALID_URL, url).groups()
 
+        page = self._download_webpage(url, video_id, 'Downloading video page')
+        preloaded_state_regex = r'window\.__PRELOADED_STATE__\s*=\s*(\{(.*?)\});'
+        preloaded_state_text = self._html_search_regex(preloaded_state_regex, page, 'preloaded state')
+        preloaded_state = self._parse_json(preloaded_state_text, video_id)
+        link_info = next(iter(preloaded_state['links'].values()))
+
         try:
-            episode = self._download_json(
-                self._API_BASE + 'client/v1/player/episode/' + video_id, video_id)
+            metadata = self._download_json(link_info['href'], video_id)
+        except ExtractorError as e:
+            self._handle_error(e, 403)
+
+        try:
+            episode = self._download_json(metadata['urlVideo'], video_id)
         except ExtractorError as e:
             self._handle_error(e, 403)
 
         title = episode['titulo']
 
         formats = []
+        subtitles = {}
         for source in episode.get('sources', []):
             src = source.get('src')
             if not src:
@@ -96,8 +132,14 @@ class AtresPlayerIE(InfoExtractor):
                     src, video_id, 'mp4', 'm3u8_native',
                     m3u8_id='hls', fatal=False))
             elif src_type == 'application/dash+xml':
-                formats.extend(self._extract_mpd_formats(
-                    src, video_id, mpd_id='dash', fatal=False))
+                mpd = self._download_xml_handle(
+                    src, video_id, note='Downloading MPD manifest', fatal=False)
+                if mpd:
+                    mpd_doc, mpd_handle = mpd
+                    mpd_base_url = base_url(mpd_handle.geturl())
+                    subtitles = self._merge_subtitles(subtitles, self._get_mpd_subtitles(mpd_doc, mpd_base_url))
+                    formats.extend(self._parse_mpd_formats(
+                        mpd_doc, mpd_id='dash', mpd_base_url=mpd_base_url, mpd_url=src))
         self._sort_formats(formats)
 
         heartbeat = episode.get('heartbeat') or {}
@@ -115,4 +157,5 @@ class AtresPlayerIE(InfoExtractor):
             'channel': get_meta('channel'),
             'season': get_meta('season'),
             'episode_number': int_or_none(get_meta('episodeNumber')),
+            'subtitles': subtitles,
         }

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)