Merge 526ef898bf1bd769f0a0f93d36312544ab0d2e1e 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/rtp.py

@@ -3,25 +3,38 @@ from __future__ import unicode_literals
 
 from .common import InfoExtractor
 from ..utils import (
+    ExtractorError,
     determine_ext,
     js_to_json,
 )
+from ..compat import (
+    compat_b64decode,
+    compat_urllib_parse_unquote,
+)
+
+import re
+
+
+def decode_b64_url(code):
+    decoded_url = re.match(r"[^[]*\[([^]]*)\]", code).groups()[0]
+    return compat_b64decode(
+        compat_urllib_parse_unquote(
+            decoded_url.replace('"', '').replace('\'', '').replace(',', ''))).decode('utf-8')
 
 
 class RTPIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?rtp\.pt/play/p(?P<program_id>[0-9]+)/(?P<id>[^/?#]+)/?'
+    _VALID_URL = r'https?://(?:(?:(?:www\.)?rtp\.pt/play/(?P<subarea>.*/)?p(?P<program_id>[0-9]+)/(?P<episode_id>e[0-9]+/)?)|(?:arquivos\.rtp\.pt/conteudos/))(?P<id>[^/?#]+)/?'
     _TESTS = [{
-        'url': 'http://www.rtp.pt/play/p405/e174042/paixoes-cruzadas',
+        'url': 'https://www.rtp.pt/play/p9165/e562949/por-do-sol',
-        'md5': 'e736ce0c665e459ddb818546220b4ef8',
         'info_dict': {
-            'id': 'e174042',
+            'id': 'por-do-sol',
-            'ext': 'mp3',
+            'ext': 'mp4',
-            'title': 'Paixões Cruzadas',
+            'title': 'Pôr do Sol Episódio 1 - de 16 Ago 2021',
-            'description': 'As paixões musicais de António Cartaxo e António Macedo',
+            'description': 'Madalena Bourbon de Linhaça vive atormentada pelo segredo que esconde desde 1990. Matilde Bourbon de Linhaça sonha fugir com o seu amor proibido. O en',
             'thumbnail': r're:^https?://.*\.jpg',
         },
     }, {
-        'url': 'http://www.rtp.pt/play/p831/a-quimica-das-coisas',
+        'url': 'https://www.rtp.pt/play/p510/aleixo-fm',
         'only_matching': True,
     }]
 
@@ -29,30 +42,61 @@ class RTPIE(InfoExtractor):
         video_id = self._match_id(url)
 
         webpage = self._download_webpage(url, video_id)
-        title = self._html_search_meta(
-            'twitter:title', webpage, display_name='title', fatal=True)
 
-        config = self._parse_json(self._search_regex(
+        # Remove JS multi-line comments from webpage source
-            r'(?s)RTPPlayer\(({.+?})\);', webpage,
+        webpage = re.sub(r'(\/\*.*\*\/)', '', webpage, flags=re.DOTALL)
-            'player config'), video_id, js_to_json)
+
-        file_url = config['file']
+        title = self._html_search_regex(r'<title>(.+?)</title>', webpage, 'title')
-        ext = determine_ext(file_url)
+
-        if ext == 'm3u8':
+        # Replace irrelevant text in title
-            file_key = config.get('fileKey')
+        title = title.replace(' - RTP Play - RTP', '')
-            formats = self._extract_m3u8_formats(
+
-                file_url, video_id, 'mp4', 'm3u8_native',
+        # Check if it's a video split in parts, if so add part number to title
-                m3u8_id='hls', fatal=file_key)
+        part = self._html_search_regex(r'section\-parts.*<span.*>(.+?)</span>.*</ul>', webpage, 'part', default=None)
-            if file_key:
+        if part:
-                formats.append({
+            title = f'{title} {part}'
-                    'url': 'https://cdn-ondemand.rtp.pt' + file_key,
+
-                    'preference': 1,
+        # Get JS object
-                })
+        js_object = self._search_regex(r'(?s)RTPPlayer *\( *({.+?}) *\);', webpage, 'player config')
+        json_string_for_config = ''
+        full_url = None
+
+        # Verify JS object since it isn't pure JSON and maybe it needs some tuning
+        for line in js_object.splitlines():
+            stripped_line = line.strip()
+
+            # key == 'fileKey', then we found what we wanted
+            if re.match(r'fileKey:', stripped_line):
+                if re.match(r'fileKey: *""', stripped_line):
+                    raise ExtractorError("Episode not found (probably removed)", expected=True)
+                url = decode_b64_url(stripped_line)
+                if 'mp3' in url:
+                    full_url = 'https://cdn-ondemand.rtp.pt' + url
+                else:
+                    full_url = 'https://streaming-vod.rtp.pt/dash{}/manifest.mpd'.format(url)
+
+            elif not stripped_line.startswith("//") and not re.match('file *:', stripped_line) and not re.match('.*extraSettings ?:', stripped_line):
+                # Ignore commented lines, `extraSettings` and `f`. The latter seems to some random unrelated video.
+                json_string_for_config += '\n' + line
+
+        if not full_url:
+            raise ExtractorError("No valid media source found in page")
+
+        # Finally send pure JSON string for JSON parsing
+        config = self._parse_json(json_string_for_config, video_id, js_to_json)
+        full_url = full_url.replace('drm-dash', 'dash')
+        ext = determine_ext(full_url)
+
+        if ext == 'mpd':
+            # Download via mpd file
+            formats = self._extract_mpd_formats(full_url, video_id)
             self._sort_formats(formats)
         else:
             formats = [{
-                'url': file_url,
+                'url': full_url,
                 'ext': ext,
             }]
+
         if config.get('mediaType') == 'audio':
             for f in formats:
                 f['vcodec'] = 'none'

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)