Merge branch 'yt-dlp:master' into master

2025-07-27 07:38:30 +00:00 · 2025-07-10 10:51:47 +03:00 · 2025-07-10 10:51:47 +03:00 · 899bf5d9b6
commit 899bf5d9b6
parent e6c1287056 2aaf1aa71d
8 changed files with 112 additions and 42 deletions
--- a/test/test_jsinterp.py
+++ b/test/test_jsinterp.py
@ -536,6 +536,11 @@ def test_nested_function_scoping(self):
            }
        ''', 31)

+    def test_undefined_varnames(self):
+        jsi = JSInterpreter('function f(){ var a; return [a, b]; }')
+        self._test(jsi, [JS_Undefined, JS_Undefined])
+        self.assertEqual(jsi._undefined_varnames, {'b'})
+

 if __name__ == '__main__':
    unittest.main()
--- a/test/test_youtube_signature.py
+++ b/test/test_youtube_signature.py
@ -373,6 +373,10 @@
        'https://www.youtube.com/s/player/e12fbea4/player_ias_tce.vflset/en_US/base.js',
        'kM5r52fugSZRAKHfo3', 'XkeRfXIPOkSwfg',
    ),
+    (
+        'https://www.youtube.com/s/player/ef259203/player_ias_tce.vflset/en_US/base.js',
+        'rPqBC01nJpqhhi2iA2U', 'hY7dbiKFT51UIA',
+    ),
 ]


--- a/yt_dlp/extractor/_extractors.py
+++ b/yt_dlp/extractor/_extractors.py
@ -1147,6 +1147,7 @@
    MindsIE,
 )
 from .minoto import MinotoIE
+from .mir24tv import Mir24TvIE
 from .mirrativ import (
    MirrativIE,
    MirrativUserIE,
--- a/yt_dlp/extractor/mir24tv.py
+++ b/yt_dlp/extractor/mir24tv.py
@ -0,0 +1,37 @@
+from .common import InfoExtractor
+from ..utils import parse_qs, url_or_none
+from ..utils.traversal import require, traverse_obj
+
+
+class Mir24TvIE(InfoExtractor):
+    IE_NAME = 'mir24.tv'
+    _VALID_URL = r'https?://(?:www\.)?mir24\.tv/news/(?P<id>[0-9]+)/[^/?#]+'
+    _TESTS = [{
+        'url': 'https://mir24.tv/news/16635210/dni-kultury-rossii-otkrylis-v-uzbekistane.-na-prazdnichnom-koncerte-vystupili-zvezdy-rossijskoj-estrada',
+        'info_dict': {
+            'id': '16635210',
+            'title': 'Дни культуры России открылись в Узбекистане. На праздничном концерте выступили звезды российской эстрады',
+            'ext': 'mp4',
+            'thumbnail': r're:https://images\.mir24\.tv/.+\.jpg',
+        },
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id, impersonate=True)
+
+        iframe_url = self._search_regex(
+            r'<iframe\b[^>]+\bsrc=["\'](https?://mir24\.tv/players/[^"\']+)',
+            webpage, 'iframe URL')
+
+        m3u8_url = traverse_obj(iframe_url, (
+            {parse_qs}, 'source', -1, {self._proto_relative_url}, {url_or_none}, {require('m3u8 URL')}))
+        formats, subtitles = self._extract_m3u8_formats_and_subtitles(m3u8_url, video_id, 'mp4', m3u8_id='hls')
+
+        return {
+            'id': video_id,
+            'title': self._og_search_title(webpage, default=None) or self._html_extract_title(webpage),
+            'thumbnail': self._og_search_thumbnail(webpage, default=None),
+            'formats': formats,
+            'subtitles': subtitles,
+        }
--- a/yt_dlp/extractor/newspicks.py
+++ b/yt_dlp/extractor/newspicks.py
@ -1,53 +1,72 @@
-import re
-
 from .common import InfoExtractor
-from ..utils import ExtractorError
+from ..utils import (
+    clean_html,
+    parse_iso8601,
+    parse_qs,
+    url_or_none,
+)
+from ..utils.traversal import require, traverse_obj


 class NewsPicksIE(InfoExtractor):
-    _VALID_URL = r'https?://newspicks\.com/movie-series/(?P<channel_id>\d+)\?movieId=(?P<id>\d+)'
-
+    _VALID_URL = r'https?://newspicks\.com/movie-series/(?P<id>[^?/#]+)'
    _TESTS = [{
-        'url': 'https://newspicks.com/movie-series/11?movieId=1813',
+        'url': 'https://newspicks.com/movie-series/11/?movieId=1813',
        'info_dict': {
            'id': '1813',
-            'title': '日本の課題を破壊せよ【ゲスト：成田悠輔】',
-            'description': 'md5:09397aad46d6ded6487ff13f138acadf',
-            'channel': 'HORIE ONE',
-            'channel_id': '11',
-            'release_date': '20220117',
-            'thumbnail': r're:https://.+jpg',
            'ext': 'mp4',
+            'title': '日本の課題を破壊せよ【ゲスト：成田悠輔】',
+            'cast': 'count:4',
+            'description': 'md5:09397aad46d6ded6487ff13f138acadf',
+            'duration': 2940,
+            'release_date': '20220117',
+            'release_timestamp': 1642424400,
+            'series': 'HORIE ONE',
+            'series_id': '11',
+            'thumbnail': r're:https?://resources\.newspicks\.com/.+\.(?:jpe?g|png)',
+            'timestamp': 1642424420,
+            'upload_date': '20220117',
+        },
+    }, {
+        'url': 'https://newspicks.com/movie-series/158/?movieId=3932',
+        'info_dict': {
+            'id': '3932',
+            'ext': 'mp4',
+            'title': '【検証】専門家は、KADOKAWAをどう見るか',
+            'cast': 'count:3',
+            'description': 'md5:2c2d4bf77484a4333ec995d676f9a91d',
+            'duration': 1320,
+            'release_date': '20240622',
+            'release_timestamp': 1719088080,
+            'series': 'NPレポート',
+            'series_id': '158',
+            'thumbnail': r're:https?://resources\.newspicks\.com/.+\.(?:jpe?g|png)',
+            'timestamp': 1719086400,
+            'upload_date': '20240622',
        },
    }]

    def _real_extract(self, url):
-        video_id, channel_id = self._match_valid_url(url).group('id', 'channel_id')
+        series_id = self._match_id(url)
+        video_id = traverse_obj(parse_qs(url), ('movieId', -1, {str}, {require('movie ID')}))
        webpage = self._download_webpage(url, video_id)
-        entries = self._parse_html5_media_entries(
-            url, webpage.replace('movie-for-pc', 'movie'), video_id, 'hls')
-        if not entries:
-            raise ExtractorError('No HTML5 media elements found')
-        info = entries[0]

-        title = self._html_search_meta('og:title', webpage, fatal=False)
-        description = self._html_search_meta(
-            ('og:description', 'twitter:title'), webpage, fatal=False)
-        channel = self._html_search_regex(
-            r'value="11".+?<div\s+class="title">(.+?)</div', webpage, 'channel name', fatal=False)
-        if not title or not channel:
-            title, channel = re.split(r'\s*|\s*', self._html_extract_title(webpage))
+        fragment = self._search_nextjs_data(webpage, video_id)['props']['pageProps']['fragment']
+        m3u8_url = traverse_obj(fragment, ('movie', 'movieUrl', {url_or_none}, {require('m3u8 URL')}))
+        formats, subtitles = self._extract_m3u8_formats_and_subtitles(m3u8_url, video_id, 'mp4')

-        release_date = self._search_regex(
-            r'<span\s+class="on-air-date">\s*(\d+)年(\d+)月(\d+)日\s*</span>',
-            webpage, 'release date', fatal=False, group=(1, 2, 3))
-
-        info.update({
+        return {
            'id': video_id,
-            'title': title,
-            'description': description,
-            'channel': channel,
-            'channel_id': channel_id,
-            'release_date': ('%04d%02d%02d' % tuple(map(int, release_date))) if release_date else None,
-        })
-        return info
+            'formats': formats,
+            'series': traverse_obj(fragment, ('series', 'title', {str})),
+            'series_id': series_id,
+            'subtitles': subtitles,
+            **traverse_obj(fragment, ('movie', {
+                'title': ('title', {str}),
+                'cast': ('relatedUsers', ..., 'displayName', {str}, filter, all, filter),
+                'description': ('explanation', {clean_html}),
+                'release_timestamp': ('onAirStartDate', {parse_iso8601}),
+                'thumbnail': (('image', 'coverImageUrl'), {url_or_none}, any),
+                'timestamp': ('published', {parse_iso8601}),
+            })),
+        }
--- a/yt_dlp/extractor/ninegag.py
+++ b/yt_dlp/extractor/ninegag.py
@ -1,6 +1,5 @@
 from .common import InfoExtractor
 from ..utils import (
-    ExtractorError,
    determine_ext,
    int_or_none,
    traverse_obj,
@ -61,10 +60,10 @@ def _real_extract(self, url):
        post = self._download_json(
            'https://9gag.com/v1/post', post_id, query={
                'id': post_id,
-            })['data']['post']
+            }, impersonate=True)['data']['post']

        if post.get('type') != 'Animated':
-            raise ExtractorError(
+            self.raise_no_formats(
                'The given url does not contain a video',
                expected=True)

--- a/yt_dlp/extractor/youtube/_video.py
+++ b/yt_dlp/extractor/youtube/_video.py
@ -3273,6 +3273,10 @@ def append_client(*client_names):
                # web_creator may work around age-verification for all videos but requires PO token
                append_client('tv_embedded', 'web_creator')

+            status = traverse_obj(pr, ('playabilityStatus', 'status', {str}))
+            if status not in ('OK', 'LIVE_STREAM_OFFLINE', 'AGE_CHECK_REQUIRED', 'AGE_VERIFICATION_REQUIRED'):
+                self.write_debug(f'{video_id}: {client} player response playability status: {status}')
+
        prs.extend(deprioritized_prs)

        if skipped_clients:
--- a/yt_dlp/jsinterp.py
+++ b/yt_dlp/jsinterp.py
@ -677,8 +677,9 @@ def dict_item(key, val):
                # Set value as JS_Undefined or its pre-existing value
                local_vars.set_local(var, ret)
            else:
-                ret = local_vars.get(var, JS_Undefined)
-                if ret is JS_Undefined:
+                ret = local_vars.get(var, NO_DEFAULT)
+                if ret is NO_DEFAULT:
+                    ret = JS_Undefined
                    self._undefined_varnames.add(var)
            return ret, should_return