Merge f43daa36fa3f9946302eadc1567084dd29cffd2b 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/extractors.py

@@ -736,6 +736,7 @@ from .mtv import (
     MTVJapanIE,
 )
 from .muenchentv import MuenchenTVIE
+from .murrtube import MurrtubeIE, MurrtubeUserIE
 from .mwave import MwaveIE, MwaveMeetGreetIE
 from .mychannels import MyChannelsIE
 from .myspace import MySpaceIE, MySpaceAlbumIE

youtube_dl/extractor/murrtube.py

@@ -0,0 +1,218 @@
+# coding: utf-8
+from __future__ import unicode_literals
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+    int_or_none,
+    try_get,
+    unified_strdate,
+)
+
+
+class MurrtubeIE(InfoExtractor):
+    _VALID_URL = r'''(?x)
+                        (?:
+                            murrtube:|
+                            https?://murrtube\.net/videos/(?:[a-z0-9\-]+)\-
+                        )
+                        (?P<id>[a-f0-9]{8}\-[a-f0-9]{4}\-[a-f0-9]{4}\-[a-f0-9]{4}\-[a-f0-9]{12})
+                    '''
+    _TEST = {
+        'url': 'https://murrtube.net/videos/inferno-x-skyler-148b6f2a-fdcc-4902-affe-9c0f41aaaca0',
+        'md5': '169f494812d9a90914b42978e73aa690',
+        'info_dict': {
+            'id': '148b6f2a-fdcc-4902-affe-9c0f41aaaca0',
+            'ext': 'mp4',
+            'title': 'Inferno X Skyler',
+            'description': 'Humping a very good slutty sheppy (roomate)',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 284,
+            'uploader': 'Inferno Wolf',
+            'upload_date': '20200502',
+            'age_limit': 18,
+        }
+    }
+
+    def _download_gql(self, video_id, op, note=None, fatal=True):
+        result = self._download_json(
+            "https://murrtube.net/graphql",
+            video_id,
+            note,
+            data=json.dumps(op).encode(),
+            headers={
+                'Content-Type': "application/json",
+            },
+            fatal=fatal,
+        )
+        return result['data']
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        data = self._download_gql(video_id, {
+            "operationName": "Medium",
+            "variables": {
+                "id": video_id,
+            },
+            "query": """query Medium($id: ID!) {
+  medium(id: $id) {
+    id
+    slug
+    title
+    description
+    key
+    duration
+    commentsCount
+    likesCount
+    liked
+    viewsCount
+    publishedAt
+    thumbnailKey
+    smallThumbnailKey
+    tagList
+    user {
+      id
+      slug
+      name
+      avatar
+      __typename
+    }
+    __typename
+  }
+}"""})
+        meta = data['medium']
+
+        storage_url = 'https://storage.murrtube.net/murrtube/'
+        format_url = storage_url + meta['key']
+        thumbnail = storage_url + meta['thumbnailKey']
+
+        formats = []
+        if determine_ext(format_url) == 'm3u8':
+            formats.extend(self._extract_m3u8_formats(
+                format_url, video_id, 'mp4', entry_protocol='m3u8_native',
+                preference=1, fatal=False))
+        else:
+            formats.append({
+                'url': format_url,
+            })
+
+        return {
+            'id': video_id,
+            'title': meta['title'],
+            'description': meta.get('description'),
+            'formats': formats,
+            'thumbnail': thumbnail,
+            'duration': int(meta['duration']),
+            'uploader': try_get(meta, lambda x: x['user']['name']),
+            'upload_date': unified_strdate(meta.get('publishedAt')),
+            'view_count': int_or_none(meta.get('viewsCount')),
+            'like_count': int_or_none(meta.get('likesCount')),
+            'comment_count': int_or_none(meta.get('commentsCount')),
+            'tags': meta.get('tagList', []),
+            'age_limit': 18,
+        }
+
+
+class MurrtubeUserIE(MurrtubeIE):
+    IE_DESC = 'Murrtube user profile'
+    _VALID_URL = r'https?://murrtube\.net/(?P<id>[^/]+)/?$'
+    _TEST = {
+        'url': 'https://murrtube.net/stormy',
+        'info_dict': {
+            'id': 'stormy',
+            'age_limit': 18,
+        },
+        'playlist_mincount': 28,
+    }
+
+    def _real_extract(self, url):
+        username = self._match_id(url)
+        data = self._download_gql(username, {
+            "operationName": "User",
+            "variables": {
+                "id": username,
+            },
+            "query": """query User($id: ID!) {
+  user(id: $id) {
+    id
+    public
+    chatEnabled
+    slug
+    name
+    avatar
+    banner
+    bio
+    website
+    followed
+    following
+    mediaCount
+    followersCount
+    followingCount
+    likesCount
+    blocked
+    __typename
+  }
+}"""},
+            "Downloading user info")
+        user = data['user']
+
+        page_size = 10
+        num_pages = int(user['mediaCount']) // page_size + 1
+        entries = []
+
+        for page in range(num_pages):
+            data = self._download_gql(username, {
+                "operationName": "Media",
+                "variables": {
+                    "limit": page_size,
+                    "offset": page * page_size,
+                    "sort": "latest",
+                    "userId": user['id'],
+                },
+                "query": """query Media($q: String, $sort: String, $userId: ID, $offset: Int!, $limit: Int!) {
+  media(q: $q, sort: $sort, userId: $userId, offset: $offset, limit: $limit) {
+    id
+    slug
+    title
+    description
+    previewKey
+    thumbnailKey
+    smallThumbnailKey
+    publishedAt
+    duration
+    commentsCount
+    likesCount
+    viewsCount
+    tagList
+    visibility
+    restriction
+    user {
+      id
+      slug
+      name
+      avatar
+      __typename
+    }
+    __typename
+  }
+  users(q: $q, fillWithFollowing: false, offset: 0, limit: 2) {
+    id
+    slug
+    name
+    avatar
+    mediaCount
+    __typename
+  }
+}"""},
+                "Downloading page {0}".format(page + 1))
+            media = data['media']
+
+            for entry in media:
+                entries.append(self.url_result(
+                    'murrtube:{0}'.format(entry['id']),
+                    MurrtubeIE.ie_key()))
+
+        playlist = self.playlist_result(entries, username)
+        playlist['age_limit'] = 18
+        return playlist

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)