Merge 5c1e5a1160a3e65fd2673b6437f3882d348991cb 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>

.gitignore

@@ -1,3 +1,4 @@
+__pycache__/
 *.pyc
 *.pyo
 *.class
@@ -5,6 +6,7 @@
 *.DS_Store
 wine-py2exe/
 py2exe.log
+.pytest_cache/
 *.kate-swp
 build/
 dist/

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

@@ -1328,7 +1328,10 @@ from .trovo import (
 from .trunews import TruNewsIE
 from .trutv import TruTVIE
 from .tube8 import Tube8IE
-from .tubitv import TubiTvIE
+from .tubitv import (
+    TubiTvIE,
+    TubiTvShowIE,
+)
 from .tumblr import TumblrIE
 from .tunein import (
     TuneInClipIE,

youtube_dl/extractor/tubitv.py

@@ -1,23 +1,62 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
+    get_element_by_id,
     int_or_none,
+    join_nonempty,
+    js_to_json,
+    merge_dicts,
+    parse_age_limit,
     sanitized_Request,
+    strip_or_none,
+    T,
+    traverse_obj,
+    url_or_none,
     urlencode_postdata,
 )
 
 
 class TubiTvIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?tubitv\.com/(?:video|movies|tv-shows)/(?P<id>[0-9]+)'
+    IE_NAME = 'tubitv'
+    _VALID_URL = r'https?://(?:www\.)?tubitv\.com/(?P<type>video|movies|tv-shows)/(?P<id>\d+)'
     _LOGIN_URL = 'http://tubitv.com/login'
     _NETRC_MACHINE = 'tubitv'
-    _GEO_COUNTRIES = ['US']
     _TESTS = [{
+        'url': 'https://tubitv.com/movies/100004539/the-39-steps',
+        'info_dict': {
+            'id': '100004539',
+            'ext': 'mp4',
+            'title': 'The 39 Steps',
+            'description': 'md5:bb2f2dd337f0dc58c06cb509943f54c8',
+            'uploader_id': 'abc2558d54505d4f0f32be94f2e7108c',
+            'release_year': 1935,
+            'thumbnail': r're:^https?://.+\.(jpe?g|png)$',
+            'duration': 5187,
+        },
+        'params': {'skip_download': 'm3u8'},
+        'skip': 'This content is currently unavailable',
+    }, {
+        'url': 'https://tubitv.com/tv-shows/554628/s01-e01-rise-of-the-snakes',
+        'info_dict': {
+            'id': '554628',
+            'ext': 'mp4',
+            'title': 'S01:E01 - Rise of the Snakes',
+            'description': 'md5:ba136f586de53af0372811e783a3f57d',
+            'episode': 'Rise of the Snakes',
+            'episode_number': 1,
+            'season': 'Season 1',
+            'season_number': 1,
+            'uploader_id': '2a9273e728c510d22aa5c57d0646810b',
+            'release_year': 2011,
+            'thumbnail': r're:^https?://.+\.(jpe?g|png)$',
+            'duration': 1376,
+        },
+        'params': {'skip_download': 'm3u8'},
+        'skip': 'This content is currently unavailable',
+    }, {
         'url': 'http://tubitv.com/video/283829/the_comedian_at_the_friday',
         'md5': '43ac06be9326f41912dc64ccf7a80320',
         'info_dict': {
@@ -27,6 +66,7 @@ class TubiTvIE(InfoExtractor):
             'description': 'A stand up comedian is forced to look at the decisions in his life while on a one week trip to the west coast.',
             'uploader_id': 'bc168bee0d18dd1cb3b86c68706ab434',
         },
+        'skip': 'Content Unavailable',
     }, {
         'url': 'http://tubitv.com/tv-shows/321886/s01_e01_on_nom_stories',
         'only_matching': True,
@@ -34,24 +74,42 @@ class TubiTvIE(InfoExtractor):
         'url': 'http://tubitv.com/movies/383676/tracker',
         'only_matching': True,
     }, {
-        'url': 'https://tubitv.com/movies/560057/penitentiary?start=true',
+        'url': 'https://tubitv.com/tv-shows/200141623/s01-e01-episode-1',
         'info_dict': {
-            'id': '560057',
+            'id': '200141623',
             'ext': 'mp4',
-            'title': 'Penitentiary',
+            'title': 'Shameless S01:E01 - Episode 1',
-            'description': 'md5:8d2fc793a93cc1575ff426fdcb8dd3f9',
+            'description': 'Having her handbag stolen proves to be a blessing in disguise for Fiona when handsome stranger Steve comes to her rescue.',
-            'uploader_id': 'd8fed30d4f24fcb22ec294421b9defc2',
+            'timestamp': 1725148800,
-            'release_year': 1979,
+            'upload_date': '20240901',
+            'uploader': 'all3-media',
+            'uploader_id': '9b8e3a8d789b1c843f4b680c025a1853',
+            'release_year': 2004,
+            'episode': 'Episode 1',
+            'episode_number': 1,
+            'season': 'Season 1',
+            'season_number': 1,
+            'series': 'Shameless',
+            'cast': list,
+            'age_limit': 17,
         },
         'params': {
-            'skip_download': True,
+            'format': 'best/bestvideo',
+            'skip_download': 'm3u8'
         },
     }]
 
+    # DRM formats are included only to raise appropriate error
+    _UNPLAYABLE_FORMATS = ('hlsv6_widevine', 'hlsv6_widevine_nonclearlead', 'hlsv6_playready_psshv0',
+                           'hlsv6_fairplay', 'dash_widevine', 'dash_widevine_nonclearlead')
+
     def _login(self):
         username, password = self._get_login_info()
         if username is None:
             return
+        self._perform_login(username, password)
+
+    def _perform_login(self, username, password):
         self.report_login()
         form_data = {
             'username': username,
@@ -62,7 +120,7 @@ class TubiTvIE(InfoExtractor):
         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
         login_page = self._download_webpage(
             request, None, False, 'Wrong login info')
-        if not re.search(r'id="tubi-logout"', login_page):
+        if get_element_by_id('tubi-logout', login_page) is None:
             raise ExtractorError(
                 'Login failed (invalid username/password)', expected=True)
 
@@ -70,41 +128,146 @@ class TubiTvIE(InfoExtractor):
         self._login()
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        video_id, video_type = self._match_valid_url(url).group('id', 'type')
-        video_data = self._download_json(
+        webpage = self._download_webpage('https://tubitv.com/{0}/{1}/'.format(video_type, video_id), video_id)
-            'http://tubitv.com/oz/videos/%s/content' % video_id, video_id)
+        data = self._search_json(
-        title = video_data['title']
+            r'window\.__data\s*=', webpage, 'data', video_id,
+            transform_source=js_to_json)['video']['byId']
+        video_data = data[video_id]
+        info = self._search_json_ld(webpage, video_id, expected_type='VideoObject', default={})
+        title = strip_or_none(info.get('title'))
+        info['title'] = title or strip_or_none(video_data['title'])
 
-        formats = self._extract_m3u8_formats(
+        formats = []
-            self._proto_relative_url(video_data['url']),
+        drm_formats = 0
-            video_id, 'mp4', 'm3u8_native')
+
+        for resource in traverse_obj(video_data, ('video_resources', lambda _, v: v['type'] and v['manifest']['url'])):
+            manifest_url = url_or_none(resource['manifest']['url'])
+            if not manifest_url:
+                continue
+            resource_type = resource['type']
+            if resource_type == 'dash':
+                formats.extend(self._extract_mpd_formats(manifest_url, video_id, mpd_id=resource_type, fatal=False))
+            elif resource_type in ('hlsv3', 'hlsv6'):
+                formats.extend(self._extract_m3u8_formats(manifest_url, video_id, 'mp4', m3u8_id=resource_type, fatal=False))
+            elif resource_type in self._UNPLAYABLE_FORMATS:
+                drm_formats += 1
+            else:
+                self.report_warning('Skipping unknown resource type "{0}"'.format(resource_type))
+
+        if not formats and drm_formats > 0:
+            self.report_drm(video_id)
+        elif not formats and not video_data.get('policy_match'):  # policy_match is False if content was removed
+            raise ExtractorError('This content is currently unavailable', expected=True)
         self._sort_formats(formats)
 
-        thumbnails = []
-        for thumbnail_url in video_data.get('thumbnails', []):
-            if not thumbnail_url:
-                continue
-            thumbnails.append({
-                'url': self._proto_relative_url(thumbnail_url),
-            })
-
         subtitles = {}
-        for sub in video_data.get('subtitles', []):
+        for sub in traverse_obj(video_data, ('subtitles', lambda _, v: v['url'])):
-            sub_url = sub.get('url')
+            sub_url = self._proto_relative_url(sub['url'])
             if not sub_url:
                 continue
             subtitles.setdefault(sub.get('lang', 'English'), []).append({
-                'url': self._proto_relative_url(sub_url),
+                'url': sub_url,
             })
 
-        return {
+        season_number, episode_number, episode_title = self._search_regex(
+            r'\bS(\d+):E(\d+) - (.+)', info['title'], 'episode info', fatal=False, group=(1, 2, 3), default=(None, None, None))
+
+        return merge_dicts({
             'id': video_id,
-            'title': title,
             'formats': formats,
             'subtitles': subtitles,
-            'thumbnails': thumbnails,
+            'season_number': int_or_none(season_number),
-            'description': video_data.get('description'),
+            'episode_number': int_or_none(episode_number),
-            'duration': int_or_none(video_data.get('duration')),
+            'episode': episode_title
-            'uploader_id': video_data.get('publisher_id'),
+        }, traverse_obj(video_data, {
-            'release_year': int_or_none(video_data.get('year')),
+            'description': ('description', T(strip_or_none)),
-        }
+            'duration': ('duration', T(int_or_none)),
+            'uploader': ('import_id', T(strip_or_none)),
+            'uploader_id': ('publisher_id', T(strip_or_none)),
+            'release_year': ('year', T(int_or_none)),
+            'thumbnails': ('thumbnails', Ellipsis, T(self._proto_relative_url), {'url': T(url_or_none)}),
+            'cast': ('actors', Ellipsis, T(strip_or_none)),
+            'categories': ('tags', Ellipsis, T(strip_or_none)),
+            'age_limit': ('ratings', 0, 'value', T(parse_age_limit)),
+        }), traverse_obj(data, (lambda _, v: v['type'] == 's', {
+            'series': ('title', T(strip_or_none)),
+            # 'series_id': ('id', T(compat_str)),
+        }), get_all=False), info)
+
+
+class TubiTvShowIE(InfoExtractor):
+    IE_NAME = 'tubitv:series'
+    _VALID_URL = r'https?://(?:www\.)?tubitv\.com/series/\d+/(?P<show_name>[^/?#]+)(?:/season-(?P<season>\d+))?'
+    _TESTS = [{
+        'url': 'https://tubitv.com/series/3936/the-joy-of-painting-with-bob-ross?start=true',
+        'playlist_mincount': 390,
+        'info_dict': {
+            'id': 'the-joy-of-painting-with-bob-ross',
+        },
+    }, {
+        'url': 'https://tubitv.com/series/3936/the-joy-of-painting-with-bob-ross/season-1',
+        'playlist_count': 13,
+        'info_dict': {
+            'id': 'the-joy-of-painting-with-bob-ross-season-1',
+        },
+    }, {
+        'url': 'https://tubitv.com/series/3936/the-joy-of-painting-with-bob-ross/season-3',
+        'playlist_count': 13,
+        'info_dict': {
+            'id': 'the-joy-of-painting-with-bob-ross-season-3',
+        },
+    }]
+
+    def _real_extract(self, url):
+        playlist_id, selected_season = self._match_valid_url(url).group(
+            'show_name', 'season')
+
+        def entries(s_url, playlist_id, selected_season_num):
+
+            def get_season_data(s_num, fatal=False):
+                if s_num is None:
+                    url, s_id = s_url, playlist_id
+                else:
+                    url = '%s/season-%d' % (s_url, s_num)
+                    s_id = '%s-season-%d' % (playlist_id, s_num)
+                webpage = self._download_webpage(url, s_id, fatal=fatal)
+                data = self._search_json(
+                    r'window\s*\.\s*__data\s*=', webpage or '', 'data', s_id,
+                    transform_source=js_to_json, default={})
+                return data['video'] if fatal else data.get('video', {})
+
+            data = get_season_data(None, fatal=True)
+            # The {series_id}.seasons JSON may lack some episodes that are available
+            # Iterate over the season numbers instead [1]
+            # 1. https://github.com/yt-dlp/yt-dlp/issues/11170#issuecomment-2399918777
+            seasons = (
+                traverse_obj(data, (
+                    'byId', lambda _, v: v['type'] == 's', 'seasons', Ellipsis,
+                    'number', T(int_or_none)))
+                if selected_season is None
+                else [selected_season])
+
+            unavail_cnt = 0
+            select_episodes = lambda _, v: v['type'] == 'v'
+            for season_number in seasons:
+                if not data:
+                    data = get_season_data(season_number)
+
+                unavail_cnt += len(traverse_obj(data, ('byId', select_episodes, 'policy_match', T(lambda m: (not m) or None))))
+
+                for episode_id, episode in traverse_obj(data, ('byId', select_episodes, T(lambda e: (e['id'], e)))):
+                    yield merge_dicts(self.url_result(
+                        'https://tubitv.com/tv-shows/{0}/'.format(episode_id), TubiTvIE.ie_key(), episode_id), {
+                            'season_number': season_number,
+                            'episode_number': int_or_none(episode.get('num')),
+                    })
+
+                data = None
+
+            if unavail_cnt > 0:
+                self.report_warning('%d items were marked as unavailable: check that the desired content is available or provide login parameters if needed' % unavail_cnt)
+
+        return self.playlist_result(
+            entries(url, playlist_id, int_or_none(selected_season)),
+            join_nonempty(playlist_id, selected_season, delim='-season-'))

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)